271 
CONCLUSIONS 
Traditional methods of producing land cover maps are expensive in both 
time and personnel. The use of digital satellite data in a semi- 
automated classification is not yet suitable for extracting details of 
a complex urban environment. A hybrid approach has been developed using 
conventional aerial photographic interpretation to obtain the required 
detail and classification accuracy and a low cost microcomputer system 
for data storage and processing. A grid cell scheme was used to 
standardize data recording and allow easy access for comparisons. 
Tests of the method on a pilot area of about 29 km 2 in the rural/urban 
fringe of Aberdeen reveal that land cover changes can be recorded, 
displayed, and measured successfully. The lineprinter maps and 
statistical tables provide adequate data for analysis. 
Comparison of 1966 and 1981 area statistics for crops and grass derived 
from this approach and from the official farm census shows very close 
agreement. 
REFERENCES 
Anderson, J.R., Hardy, E.E., Roach, J.T. and Witmer, R.E. 1976, A Land 
Use and Land Cover Classification System for Use with Remote Sensor 
Data: U.S. Geological Survey Professional Paper 964, 28 pp. 
Carter, P. 1979, Urban Monitoring with LANDSAT Imagery: Machine Aided 
Image Analysis, Institute of Physics, Bristol, pp. 220-228. 
Coleman, A. 1961, The Second Land Use Survey: Geographical Journal, Vol. 
127, No. 2, pp. 168-186. 
Stamp, L.D. 1931, The Land Utilisation of Britain: Geographical Journal, 
Vol. 78, No. 1, pp. 40-53. 
Stove, G.C. et al. 1981, An Experimental Land Use Map for Regional 
Planning Based on Interpretation and Classification of Multi-temporal 
LANDSAT Data: Remote Sensing in Geological and Terrain Studies, Remote 
Sensing Society, London. 
Van Genderen, J.L. and Smith, T.F. 1976, A National Land Use Survey of 
the Developed Areas of England and Wales by Remote Sensing: Proceedings 
of the 10th International Symposium on Remote Sensing of Environment, 
Ann Arbor, Vol. 1, pp. 383-392.